Simulating the effects of pesticides on honey bee (Apis mellifera L.) colonies with BeePop+

Authors: GARBER, K. - Environmental Protection Agency (EPA); DeGrandi-Hoffman, Gloria; CURRY, R. - Crystal River Consulting Llc; MINUCCI, J.M. - Us Environmental Protection Agency (EPA); DAWSON, D. - Environmental Protection Agency (EPA); DOUGLASS, C. - US Department Of Agriculture (USDA); MILONE, J.P. - Us Environmental Protection Agency (EPA); PURUCKER, S.T. - Us Environmental Protection Agency (EPA)

Submitted to: Ecologies
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 7/6/2022
Publication Date: 7/25/2022
Citation: Garber, K., Hoffman, G.D., Curry, R., Minucci, J., Dawson, D.E., Douglass, C., Milone, J., Purucker, S.T. 2022. Simulating the effects of pesticides on honey bee (Apis mellifera L.) colonies with BeePop+. Ecologies. 3(3):275-291. https://doi.org/10.3390/ecologies3030022.
DOI: https://doi.org/10.3390/ecologies3030022

Interpretive Summary: When assessing risks that pesticides might pose to non-target organisms such a honey bees, the US-Environmental Protection Agency (USEPA) employs a tiered process. If risks are identified based on direct exposure and toxicity data for individual adult and larval honey bees (tier-1), the effects are evaluated in higher-tier studies using honey bee colonies. Colony-level studies are expensive however, and can produce data that are difficult to interpret due to the inherent variability of colonies and factors that cannot readily be controlled (i.e., weather, pests, diseases, available forage, and bee management practices). To better interpret data from colony risk assessment studies, the USEPA and the US Department of Agriculture (USDA) developed a simulation model, BeePop+, that assesses potential honey bee colony-level effects of pesticides. In this manuscript, we describe BeePop+ using the Population Model Guidance, Use, Interpretation, and Development for Ecological risk assessment (Pop-GUIDE) framework. Pop-GUIDE is a conceptual framework for the development and evaluation of population models. Within the context of Pop-GUIDE, BeePop+ is considered a “realistic-precise” model and reflects the inherent variability of colony response to pesticide exposure by simulating many outcomes. BeePop+ meets the desired features needed for use in pesticide risk assessments as its required data inputs are typically available, it is applicable to different US locations, and the outputs are both relevant to USEPA’s protection goals for honey bees and are consistent with the outcomes of empirical studies. The model also has been evaluated using available empirical colony-level data; however additional evaluation with other studies may still be done in the future prior to completing implementation.

Technical Abstract: The US Environmental Protection Agency (USEPA) employs a tiered process for assessing risks of pesticides to bees. The model discussed in this paper focuses on honey bees (Apis mellifera L.). If risks to honey bees are identified at the first tier based on exposure and toxicity data for individual adult and larval honey bees, then effects are evaluated in higher-tier studies using honey bee colonies. Colony-level studies require large amounts of resources (to conduct and review) and can yield data complicated by the inherent variability of colonies, which are influenced by factors that cannot readily be controlled, including weather, pests, diseases, available forage, and bee management practices. To better interpret these data, the USEPA and the US Department of Agriculture (USDA) developed a simulation model, BeePop+, that assesses potential honey bee colony-level effects of pesticides. Here we describe this model using the Population model Guidance, Use, Interpretation, and Development for Ecological risk assessment (Pop-GUIDE) framework, which is a conceptual framework for the development and evaluation of population models. Within the context of Pop-GUIDE, BeePop+ is considered a “realistic-precise” model and reflects the inherent variability of colony response to pesticide exposure by simulating many outcomes. This model meets the desired features needed for use in pesticide risk assessments as its required data inputs are typically available, it is applicable to different US locations, and the outputs are both relevant to USEPA’s protection goals for honey bees and are consistent with the outcomes of empirical studies. This model has also been evaluated using available empirical colony-level data; however additional evaluation with other studies may still be done in the future prior to completing implementation.